Koreen L.

Key Skills

Software

Data Annotation Tech

Labelbox

Mercor

Axiom AI

Top Subject Matter

Education

Social Science

Theology

Top Data Types

Image

Video

Text

Top Task Types

Bounding Box

Classification

Text Generation

Question Answering

Text Summarization

Fine-tuning

Data Collection

Freelancer Overview

I evaluated LLMs based on their responses to user prompts across specific categories. I worked on a variety of projects regarding images, videos, and text output. I researched generic scenarios, provided documents to support the claim, and created a prompt to confuse the LLM. I analyzed the mistake. I provided eclectic evidence for rating responses against each other.

Labeling Experience

Retrieval-Augmented Generation

DocumentText Generation

I worked on three of these types of projects. Each came with guidelines - to create a prompt that would cause the LLM to fail. Each required uploading three or more documents, as well as links to support my research. When the LLM failed, I was required to write an evaluation to pinpoint the discrepancy. One particular project was to tell the LLM that it was an administrator meeting with an autistic student and his family, and to create a PDF to help the student transition from high school into college.

2026 - 2026

Education

Work History